When AI Becomes Less Intelligent Beyond the English-speaking World
In an era dominated by intelligent automation, where industrial automation and automation equipment drive efficiency across sectors, the reliance of social media on English-language AI models for content moderation reveals a critical blind spot. While AI technologies promise to scale digital safety through automated systems, their effectiveness falters dramatically in non-English contexts. This gap underscores a fundamental flaw in how intelligent automation is deployed globally: when AI is trained primarily on English data, its “intelligence” diminishes in linguistically and culturally diverse regions, exposing marginalized communities to harm and perpetuating systemic inequities.
The English-Centric Pitfall in AI Moderation
As artificial intelligence (AI) becomes more embedded in content moderation on social media platforms, its role in maintaining digital safety has never been more important or more contentious. While AI tools have significantly improved the detection of harmful or illegal content at scale, they remain deeply flawed, especially when operating in languages other than English. Beyond the English-speaking world, the challenges of deploying AI for content moderation are not merely technical, but also cultural, political, and ethical — a failure to train AI models adequately could potentially result in harm in the real world.
Content moderation algorithms are primarily trained on English-language data. About 95 per cent of the Large Language Models (LLMs) are trained primarily on English or a combination of English with other dominant languages. A report by the Center for Democracy and Technology in 2024 similarly found that major tech companies like Meta (formerly Facebook), YouTube, and X (formerly Twitter) spent disproportionately more on English-language moderation than on other languages, even though the majority of their user base lived outside the anglophone world. As a result, AI systems often fail to understand context, nuance, or regional dialects in languages, leading to over-censorship of innocuous posts and the under-enforcement of genuinely harmful content.
A stark example is Myanmar, where Facebook faced global criticism for its role in amplifying hate speech during the Rohingya crisis of 2017. Despite Facebook’s dominance in the country, the platform was slow to recognise and act on hate speech written in Burmese. Scholars, reporters, and United Nations special rapporteurs have unanimously concluded that Facebook had been used to incite violence and that its content moderation policies were inadequate amid the genocide of Rohingya in that year. The platform has since tried to improve its Burmese language moderation, but the damage had already been done. This underscores the dangers of relying on English-centric AI systems in multilingual societies.
The Limits of Automation in Diverse Contexts
Even when local language support exists, AI struggles with local context. Sarcasm, coded language, and regional slang often slip past filters or get flagged erroneously. For example, during political protests in Thailand, protesters adopted creative and indirect ways of criticising the monarchy, such as using food metaphors or fictional characters. Lacking cultural fluency, AI systems missed these workarounds entirely or mistakenly flagged unrelated content.
The limitations of automation equipment (in this case, AI moderation tools) become evident in such scenarios. Keyword-based filtering and tokenization — common techniques in content moderation — fail when malicious actors use symbol substitution or language mixing, as seen in a recent Thai recruitment post for illicit “grey” businesses. The post, which combined Thai, English, and symbols, was obvious to native speakers but invisible to AI models, highlighting how intelligent automation reliant on English datasets cannot decode context-specific threats.
Understandably, there is also the challenge of data scarcity. Many languages lack the vast annotated datasets required to train effective AI models. This is especially true for low-resource languages spoken in parts of Africa, Southeast Asia, and Latin America. In these cases, companies rely on poorly translated datasets for machine-learning models, resulting in higher error rates.
Toward Inclusive Automation: Solutions Beyond Algorithms
To address these gaps, the industry must rethink how industrial automation and intelligent automation are applied to global content moderation. Key steps include:
- Equitable Investment in Local Languages: Tech companies must allocate resources to develop high-quality datasets for non-English languages, partnering with local communities to ensure cultural accuracy. This requires hiring and fairly compensating native-speaking moderators, even as many firms downsize such teams — a counter-trend but a necessary one for marginalized users’ safety.
- Transparency in Automation Systems: Companies should disclose which languages their AI tools support, the accuracy rates of moderation decisions, and the human oversight processes. Without clarity on how automation equipmentoperates in diverse contexts, users cannot trust the fairness of content decisions.
- Hybrid Models of AI and Human Expertise: AI alone cannot solve cultural and linguistic complexity. Integrating local moderators into intelligent automationworkflows — to train models, review edge cases, and refine algorithms — is essential. This hybrid approach ensures that AI enhances human capabilities rather than replacing them.
Conclusion: Rethinking Automation’s Global Promise
The current English-centric model of AI content moderation is a stark reminder that intelligent automation is not neutral — it reflects the biases and priorities of its creators. In a world where most internet users communicate in languages other than English, relying on Anglocentric AI systems perpetuates harm, from amplifying hate speech to enabling scams.
For industrial automation and automation equipment to serve global communities equitably, tech leaders must embrace a paradigm shift: prioritizing inclusive data, cultural expertise, and transparent governance over algorithmic speed. Until then, the promise of safe digital spaces will remain inaccessible to billions, exposing the dark underbelly of a tech-driven world that values efficiency over equity. The future of AI moderation lies not in smarter tools alone, but in recognizing that true intelligence lies in understanding — and respecting — the diversity of human expression.
